Many industrial gases contain H2S and COS. Examples include, but are not limited to fuel gases, Claus plant tail gases, and hydrocarbon feeds for reforming and other processes.
One such fuel gas, syngas, is prepared by reforming a carbonaceous fuel by contacting it with an oxygen donor under high temperature conditions to produce a fuel gas containing H2 and CO fuel components, which are typically recovered as a mixture with CO2, steam and gaseous contaminants including H2S, and COS. The carbonaceous fuel can be any of various solid, liquid, or gaseous materials having a substantial elemental content of carbon and hydrogen. Such materials include, for example, coal or coke, liquid feedstocks such as heavy naphtha fractions, and/or gaseous feedstocks such as natural gas. Commercial syngas processes typically include a desulfurization unit to remove H2S and COS sulfur species from the syngas.
Various desulfurization processes are known in the art. The current commercial process for removing H2S from steam-containing syngas streams involves cooling the initial product gas to a temperature below its dew point to remove water and then contacting the gas with an aqueous solvent containing amines. However, cooling of a fuel gas stream, such as syngas, reduces the thermal efficiency of the process often making this processing technology less advantageous compared to other competing technologies. Amine-based scrubbing processes also have technical problems such as the formation of heat stable salts, decomposition of amines, and are additionally equipment-intensive, thus requiring substantial capital investment.
In recent years, substantial research and investment has been directed towards various syngas processes, such as the “Integrated-Gasification-Combined-Cycle” (IGCC) gasification process, for generating syngas which can be used as the feed in a power plant for the generation of energy, raw material for generation of high-value chemical or transportation fuels, and a hydrogen source for fuel cells. Although this technology offers considerable improvement in both thermal and environmental efficiency, the cost of this technology is currently impeding market penetration of this technology. One approach being investigated to substantially reduce cost involves the incorporation of a water quench in the gasification process. This water quench readily removes almost all of the solid and chemical contaminants in the syngas. Unfortunately, the treatment does not remove the sulfur, and increases the steam to 60 volume percent, or more. Under these conditions, a hot-gas desulfurization process operating between 204-370° C. (400-700° F.) would have significant technical and cost advantages over other desulfurization technologies, particularly amine-based processes. Economic evaluation also indicates that this syngas process has a cost advantage over competing technologies.
The use of solid sorbents has been proposed to remove H2S and COS from power plant fuel gasses and to increase efficiency of the power plants. Preferred sorbents are regenerable materials that can be recycled and reused for numerous cycles, thereby reducing the overall process cost. Various solid sorbent materials have been used commercially to remove H2S from hydrocarbon streams. For example, zinc oxide is used in guard beds to remove H2S according to the reaction:ZnO+H2S→ZnS+H2OAlthough zinc oxide can theoretically be regenerated by burning off the sulfur at elevated temperatures according to the reaction:ZnS+3/2O2→ZnO+SO2,special sorbent compositions are necessary so that structural and chemical stability are maintained by the sorbent during regeneration. For example, zinc oxide guard bed materials are designed to have high sulfur removal activity resulting from the high surface area and zinc oxide content. However, upon regeneration the guard bed materials are destroyed, because of high temperatures required for regeneration, physical transformations caused by regeneration and competing reactions. During regeneration, the conversion of the ZnS back into ZnO results in not only the obvious chemical transformation, but also in a physical transformation associated with the size and shape of the molecules and their crystallites. The restructuring necessary for these transformations stretches, bends and twists the material altering its structure and mechanical integrity. The more ZnO converted to ZnS, the more substantial the restructuring changes are. The high temperature and exothermic nature of the regeneration reaction also increase the thermal stress and potential for sintering experienced by the material during regeneration. Finally, competing reactions result in the conversion of the ZnS into an inactive sulfate rather than ZnO.
Special sorbent compositions are not only necessary for the chemistry associated with regenerable desulfurization, but also the physical requirements of the reactor system used. One of the most promising reactor systems for this application involves transport reactor systems. These systems provide a convenient means of continuously moving sorbent material between the desulfurization reactor and regeneration reactor. These systems also provide excellent temperature control for the exothermic regeneration reactions.
Hot gas desulfurization using coupled, fluidized transport bed reactors wherein the contaminated gas stream is contacted with a solid sorbent in the first fluidized bed reactor, and the sorbent is regenerated in the second reactor, is described, for example, in Campbell, William N. and Henningsen, Gunnar B., Hot Gas Desulfurization Using Transport Reactors, publication from the M. W. Kellogg Company, pp 1059-64, 12th Annual International Pittsburgh Coal Conference Proceedings, September 1995, and in U.S. Pat. No. 5,447,702, issued on Sep. 5, 1995 to Campbell et al. Such fluidized bed processes provide substantial benefits. However, the use of fluidized beds requires that the sorbent be made in particulate form (typically 100 μm average size) and have high mechanical and chemical attrition resistance in addition to high reactivity for H2S and COS.
U.S. Pat. No. 4,088,736 to Courty et al. teaches the production of regenerable H2S sorbents from a physical mixture of zinc oxide, alumina and a Group IIA metal oxide such as calcium oxide. The mixture is calcined at elevated temperatures (>500° C.) to provide a composition in which the Group IIA metal oxide combines with alumina to yield a Group IIA metal aluminate, which is said to enhance strength properties of the final sorbent. The material is formed into pellets for use in fixed-beds.
U.S. Pat. Nos. 5,254,516 and 5,714,431 to Gupta et al. disclose processes for preparing and using sorbents based on zinc titanate. As discussed in the Gupta et al patents, particularly the '516 patent, zinc oxide-based sulfur sorbents are generally unsatisfactory for removal of sulfur from reducing gasses, such as fuel gases, at temperatures exceeding about 900° F. (482° C.) because H2 and CO components present in these gases reduce ZnO to Zn metal resulting in loss of the active zinc component by evaporation or sublimation. The zinc titanate sorbents disclosed in the Gupta et al patents are prepared by blending zinc oxide and titanium dioxide with binders followed by granulation and/or spray drying and calcinations at high temperature (750-950° C.). The resultant sorbents typically have a surface area in the range of from 0.3 to 4 m2/g, and are suitable for removal of H2S and COS contaminants from a fuel gas at high temperatures (typically in excess of 500° C.). However, these sorbents cannot, as a practical matter, be used to remove H2S and COS contaminants from fuel gas streams at temperatures in the range of 204-370° C., due to the low reactivity of the sorbents at these temperatures.
U.S. Pat. Nos. 5,494,880, 5,703,003, and 5,866,503, to Siriwardane disclose regenerable sorbent materials which include a binder, an active material and an inert material. A preferred active sorbent material is zinc oxide. These sorbent materials have numerous desirable chemical and physical properties, including the provision of a regenerable sorbent based on zinc oxide. Nevertheless, these sorbents were typically prepared in the form of 3-4 mm ellipsoidal pellets, and attempts to produce these sorbent materials for use in fluidized-bed reactors have not resulted in any material having sufficient attrition resistance to allow use in fluidized bed reactors.
PCT Application WO 99/42201 discloses regenerable, attrition resistant, spray dried, sulfur sorbents comprising a zinc titanate component and a metal aluminate, preferably zinc aluminate, component. The sorbents are preferably free of unreacted alumina. The zinc aluminate component enhances the strength, particularly the attrition resistance, of the zinc titanate. “Poisoning” of the zinc titanate by alumina, as would normally occur during high temperature regeneration of conventional zinc titanate/alumina sorbents, is substantially eliminated because the zinc-reactive sites on the alumina component are already occupied by zinc ions. Because these sorbents are based on zinc titanate, and also as a result of their high zinc aluminate content, they are only useful for substantial sulfur removal at temperatures exceeding about 500° C. (932° F.).
Numerous other sorbent materials have been proposed for removal of reduced sulfur species, such as H2S and COS, from reducing gas streams. However no currently available sorbent material possesses the necessary attributes for removal of H2S and COS contaminants from a low temperature gas stream in a system using fluidized bed desulfurization and regeneration zones due to one or more of the following deficiencies: (i) the sorbent is not regenerable; (ii) the sorbent lacks sufficient H2S and COS reactivity at temperatures below 370° C. (about 700° F.); (iii) the sorbent is not available in a physical form of a size, shape, and density, suitable for fluidization, and/or; (iv) the sorbent is not sufficiently attrition resistant for use in fluidized-bed applications.